Content
92%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body is a strong example of a lean, single-purpose skill: copy-paste-ready commands, a precise field map, an explicit fallback and retry path, and non-obvious provider-specific details that Claude would not know. Its only weakness is minor redundancy among the curl examples and format-code listings.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is efficient and assumes competence — wttr.in-specific quirks (j2 omitting hourly data, browser-oriented HTML, the output cap) are non-obvious and earn their tokens — but the five near-duplicate curl example lines plus the separate format-code list could be trimmed, fitting anchor 4 rather than 5. | 4 / 5 |
Actionability | Fully executable, copy-paste-ready guidance: a complete web_fetch call with parameters, curl commands with all flags (--fail --silent --show-error --max-time 20), an explicit JSON field map, and format codes — covering the common cases per anchor 5. | 5 / 5 |
Workflow Clarity | The single task is unambiguous with a clear decision structure: preferred tool (web_fetch), explicit fallback condition ("Use curl only if web_fetch is unavailable or disabled"), which fields to summarize, and a retry path (wttr.is); no destructive or batch operations, so the simple-skill exception applies. | 5 / 5 |
Progressive Disclosure | A compact single-file skill with no bundle files and no need for external references: everything (field map, format codes, fallbacks) is appropriately inline under clear sections (Preferred, Fallback, Notes) with nothing buried or nested — matching anchor 5's well-organized structure. | 5 / 5 |
Total | 19 / 20 Passed |